ACL RD-TEC 1.0 Summarization of C04-1147
Paper Title:
FAST COMPUTATION OF LEXICAL AFFINITY MODELS
FAST COMPUTATION OF LEXICAL AFFINITY MODELS
Authors: Egidio Terra and Charles L. A. Clarke
Primarily assigned technology terms:
- algorithm
- approximation
- binary search
- computing
- estimation procedure
- estimator
- human language
- language modeling
- length normalization
- likelihood estimator
- machine translation
- maximum likelihood
- maximum likelihood estimator
- modeling
- normalization
- parsing
- query expansion
- ranking
- ratio test
- search
- search engine
- search engines
- search system
- segmentation
- smoothing
- term weighting
- topic segmentation
- validation
- weighting
- word bigram
Other assigned terms:
- approach
- bias
- bigram
- bigram language model
- cache
- case
- co-occurrence
- co-occurrence frequency
- co-occurrences
- collocation
- content words
- contextual information
- corpora
- data structure
- data structures
- dictionary
- disk
- distribution
- document
- document boundary
- document information
- document length
- estimation
- exponential distribution
- foreign language
- geometric distribution
- heuristics
- hypothesis
- implementation
- independence assumption
- independence model
- index
- information measure
- information need
- knowledge
- language model
- language models
- large corpora
- large corpus
- latent semantic
- likelihood
- log-likelihood
- log-likelihood ratio
- measure
- measures
- mutual information
- n-gram
- n-gram model
- natural language
- pairs of words
- pointwise mutual information
- probabilities
- probability
- probability distribution
- procedure
- queries
- query
- relative frequency
- scalability
- seed
- semantic
- sentence
- similarity between words
- similarity measure
- similarity measures
- size of the corpus
- statistical models
- statistics
- synonym
- synonyms
- synonymy
- syntactic function
- target word
- technologies
- term
- term co-occurrence
- term distribution
- terms
- test set
- text
- theories
- thesaurus
- time complexity
- training
- training data
- verb
- vocabulary
- web pages
- word
- word association
- word co-occurrence
- word pair
- word similarity
- words